mediagoblin/db/models.py

   1 # GNU MediaGoblin -- federated, autonomous media hosting
   2 # Copyright (C) 2011 Free Software Foundation, Inc
   3 #
   4 # This program is free software: you can redistribute it and/or modify
   5 # it under the terms of the GNU Affero General Public License as published by
   6 # the Free Software Foundation, either version 3 of the License, or
   7 # (at your option) any later version.
   8 #
   9 # This program is distributed in the hope that it will be useful,
  10 # but WITHOUT ANY WARRANTY; without even the implied warranty of
  11 # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  12 # GNU Affero General Public License for more details.
  13 #
  14 # You should have received a copy of the GNU Affero General Public License
  15 # along with this program.  If not, see <http://www.gnu.org/licenses/>.
  16
  17 import datetime, uuid
  18
  19 from mongokit import Document
  20
  21 from mediagoblin import util
  22 from mediagoblin.auth import lib as auth_lib
  23 from mediagoblin import mg_globals
  24 from mediagoblin.db import migrations
  25 from mediagoblin.db.util import ASCENDING, DESCENDING, ObjectId
  26 from mediagoblin.util import Pagination
  27 from mediagoblin.util import DISPLAY_IMAGE_FETCHING_ORDER
  28
  29
  30 ###################
  31 # Custom validators
  32 ###################
  33
  34 ########
  35 # Models
  36 ########
  37
  38
  39 class User(Document):
  40     """
  41     A user of MediaGoblin.
  42
  43     Structure:
  44      - username: The username of this user, should be unique to this instance.
  45      - email: Email address of this user
  46      - created: When the user was created
  47      - plugin_data: a mapping of extra plugin information for this User.
  48        Nothing uses this yet as we don't have plugins, but someday we
  49        might... :)
  50      - pw_hash: Hashed version of user's password.
  51      - email_verified: Whether or not the user has verified their email or not.
  52        Most parts of the site are disabled for users who haven't yet.
  53      - status: whether or not the user is active, etc.  Currently only has two
  54        values, 'needs_email_verification' or 'active'.  (In the future, maybe
  55        we'll change this to a boolean with a key of 'active' and have a
  56        separate field for a reason the user's been disabled if that's
  57        appropriate... email_verified is already separate, after all.)
  58      - verification_key: If the user is awaiting email verification, the user
  59        will have to provide this key (which will be encoded in the presented
  60        URL) in order to confirm their email as active.
  61      - is_admin: Whether or not this user is an administrator or not.
  62      - url: this user's personal webpage/website, if appropriate.
  63      - bio: biography of this user (plaintext, in markdown)
  64      - bio_html: biography of the user converted to proper HTML.
  65     """
  66     __collection__ = 'users'
  67
  68     structure = {
  69         'username': unicode,
  70         'email': unicode,
  71         'created': datetime.datetime,
  72         'plugin_data': dict, # plugins can dump stuff here.
  73         'pw_hash': unicode,
  74         'email_verified': bool,
  75         'status': unicode,
  76         'verification_key': unicode,
  77         'is_admin': bool,
  78         'url' : unicode,
  79         'bio' : unicode,     # May contain markdown
  80         'bio_html': unicode, # May contain plaintext, or HTML
  81         }
  82
  83     required_fields = ['username', 'created', 'pw_hash', 'email']
  84
  85     default_values = {
  86         'created': datetime.datetime.utcnow,
  87         'email_verified': False,
  88         'status': u'needs_email_verification',
  89         'verification_key': lambda: unicode(uuid.uuid4()),
  90         'is_admin': False}
  91
  92     def check_login(self, password):
  93         """
  94         See if a user can login with this password
  95         """
  96         return auth_lib.bcrypt_check_password(
  97             password, self['pw_hash'])
  98
  99
 100 class MediaEntry(Document):
 101     """
 102     Record of a piece of media.
 103
 104     Structure:
 105      - uploader: A reference to a User who uploaded this.
 106
 107      - title: Title of this work
 108
 109      - slug: A normalized "slug" which can be used as part of a URL to retrieve
 110        this work, such as 'my-works-name-in-slug-form' may be viewable by
 111        'http://mg.example.org/u/username/m/my-works-name-in-slug-form/'
 112        Note that since URLs are constructed this way, slugs must be unique
 113        per-uploader.  (An index is provided to enforce that but code should be
 114        written on the python side to ensure this as well.)
 115
 116      - created: Date and time of when this piece of work was uploaded.
 117
 118      - description: Uploader-set description of this work.  This can be marked
 119        up with MarkDown for slight fanciness (links, boldness, italics,
 120        paragraphs...)
 121
 122      - description_html: Rendered version of the description, run through
 123        Markdown and cleaned with our cleaning tool.
 124
 125      - media_type: What type of media is this?  Currently we only support
 126        'image' ;)
 127
 128      - media_data: Extra information that's media-format-dependent.
 129        For example, images might contain some EXIF data that's not appropriate
 130        to other formats.  You might store it like:
 131
 132          mediaentry['media_data']['exif'] = {
 133              'manufacturer': 'CASIO',
 134              'model': 'QV-4000',
 135              'exposure_time': .659}
 136
 137        Alternately for video you might store:
 138
 139          # play length in seconds
 140          mediaentry['media_data']['play_length'] = 340
 141
 142        ... so what's appropriate here really depends on the media type.
 143
 144      - plugin_data: a mapping of extra plugin information for this User.
 145        Nothing uses this yet as we don't have plugins, but someday we
 146        might... :)
 147
 148      - tags: A list of tags.  Each tag is stored as a dictionary that has a key
 149        for the actual name and the normalized name-as-slug, so ultimately this
 150        looks like:
 151          [{'name': 'Gully Gardens',
 152            'slug': 'gully-gardens'},
 153           {'name': 'Castle Adventure Time?!",
 154            'slug': 'castle-adventure-time'}]
 155
 156      - state: What's the state of this file?  Active, inactive, disabled, etc...
 157        But really for now there are only two states:
 158         "unprocessed": uploaded but needs to go through processing for display
 159         "processed": processed and able to be displayed
 160
 161      - queued_media_file: storage interface style filepath describing a file
 162        queued for processing.  This is stored in the mg_globals.queue_store
 163        storage system.
 164
 165      - queued_task_id: celery task id.  Use this to fetch the task state.
 166
 167      - media_files: Files relevant to this that have actually been processed
 168        and are available for various types of display.  Stored like:
 169          {'thumb': ['dir1', 'dir2', 'pic.png'}
 170
 171      - attachment_files: A list of "attachment" files, ones that aren't
 172        critical to this piece of media but may be usefully relevant to people
 173        viewing the work.  (currently unused.)
 174     """
 175     __collection__ = 'media_entries'
 176
 177     structure = {
 178         'uploader': ObjectId,
 179         'title': unicode,
 180         'slug': unicode,
 181         'created': datetime.datetime,
 182         'description': unicode, # May contain markdown/up
 183         'description_html': unicode, # May contain plaintext, or HTML
 184         'media_type': unicode,
 185         'media_data': dict, # extra data relevant to this media_type
 186         'plugin_data': dict, # plugins can dump stuff here.
 187         'tags': [dict],
 188         'state': unicode,
 189
 190         # For now let's assume there can only be one main file queued
 191         # at a time
 192         'queued_media_file': [unicode],
 193         'queued_task_id': unicode,
 194
 195         # A dictionary of logical names to filepaths
 196         'media_files': dict,
 197
 198         # The following should be lists of lists, in appropriate file
 199         # record form
 200         'attachment_files': list}
 201
 202     required_fields = [
 203         'uploader', 'created', 'media_type', 'slug']
 204
 205     default_values = {
 206         'created': datetime.datetime.utcnow,
 207         'state': u'unprocessed'}
 208
 209     def get_comments(self):
 210         return self.db.MediaComment.find({
 211                 'media_entry': self['_id']}).sort('created', DESCENDING)
 212
 213     def get_display_media(self, media_map, fetch_order=DISPLAY_IMAGE_FETCHING_ORDER):
 214         """
 215         Find the best media for display.
 216
 217         Args:
 218         - media_map: a dict like
 219           {u'image_size': [u'dir1', u'dir2', u'image.jpg']}
 220         - fetch_order: the order we should try fetching images in
 221
 222         Returns:
 223         (media_size, media_path)
 224         """
 225         media_sizes = media_map.keys()
 226
 227         for media_size in DISPLAY_IMAGE_FETCHING_ORDER:
 228             if media_size in media_sizes:
 229                 return media_map[media_size]
 230
 231     def main_mediafile(self):
 232         pass
 233
 234     def generate_slug(self):
 235         self['slug'] = util.slugify(self['title'])
 236
 237         duplicate = mg_globals.database.media_entries.find_one(
 238             {'slug': self['slug']})
 239
 240         if duplicate:
 241             self['slug'] = "%s-%s" % (self['_id'], self['slug'])
 242
 243     def url_for_self(self, urlgen):
 244         """
 245         Generate an appropriate url for ourselves
 246
 247         Use a slug if we have one, else use our '_id'.
 248         """
 249         uploader = self.uploader()
 250
 251         if self.get('slug'):
 252             return urlgen(
 253                 'mediagoblin.user_pages.media_home',
 254                 user=uploader['username'],
 255                 media=self['slug'])
 256         else:
 257             return urlgen(
 258                 'mediagoblin.user_pages.media_home',
 259                 user=uploader['username'],
 260                 media=unicode(self['_id']))
 261
 262     def url_to_prev(self, urlgen):
 263         """
 264         Provide a url to the previous entry from this user, if there is one
 265         """
 266         cursor = self.db.MediaEntry.find({'_id' : {"$gt": self['_id']},
 267                                           'uploader': self['uploader'],
 268                                           'state': 'processed'}).sort(
 269                                                     '_id', ASCENDING).limit(1)
 270         if cursor.count():
 271             return urlgen('mediagoblin.user_pages.media_home',
 272                           user=self.uploader()['username'],
 273                           media=unicode(cursor[0]['slug']))
 274
 275     def url_to_next(self, urlgen):
 276         """
 277         Provide a url to the next entry from this user, if there is one
 278         """
 279         cursor = self.db.MediaEntry.find({'_id' : {"$lt": self['_id']},
 280                                           'uploader': self['uploader'],
 281                                           'state': 'processed'}).sort(
 282                                                     '_id', DESCENDING).limit(1)
 283
 284         if cursor.count():
 285             return urlgen('mediagoblin.user_pages.media_home',
 286                           user=self.uploader()['username'],
 287                           media=unicode(cursor[0]['slug']))
 288
 289     def uploader(self):
 290         return self.db.User.find_one({'_id': self['uploader']})
 291
 292
 293 class MediaComment(Document):
 294     """
 295     A comment on a MediaEntry.
 296
 297     Structure:
 298      - media_entry: The media entry this comment is attached to
 299      - author: user who posted this comment
 300      - created: when the comment was created
 301      - content: plaintext (but markdown'able) version of the comment's content.
 302      - content_html: the actual html-rendered version of the comment displayed.
 303        Run through Markdown and the HTML cleaner.
 304     """
 305
 306     __collection__ = 'media_comments'
 307
 308     structure = {
 309         'media_entry': ObjectId,
 310         'author': ObjectId,
 311         'created': datetime.datetime,
 312         'content': unicode,
 313         'content_html': unicode}
 314
 315     required_fields = [
 316         'media_entry', 'author', 'created', 'content']
 317
 318     default_values = {
 319         'created': datetime.datetime.utcnow}
 320
 321     def media_entry(self):
 322         return self.db.MediaEntry.find_one({'_id': self['media_entry']})
 323
 324     def author(self):
 325         return self.db.User.find_one({'_id': self['author']})
 326
 327
 328 REGISTER_MODELS = [
 329     MediaEntry,
 330     User,
 331     MediaComment]
 332
 333
 334 def register_models(connection):
 335     """
 336     Register all models in REGISTER_MODELS with this connection.
 337     """
 338     connection.register(REGISTER_MODELS)
 339